Day after day, I’m hearing (and seeing) a common refrain in Seer halls, at conferences, and on Twitter: we love data! The more you have, the better. Keep the new data sources coming.
I don’t completely disagree with this (it’s all about big data after all), but I think that by asking for more and more data, we’re ignoring opportunities to “recycle” data we already have to find new and different insights. Think about it–if you’re working at a full-service agency, your overall team may have access to a client’s Google Analytics and Adwords, SEMRush, STAT, HotJar, SurveyMonkey, SpyFu, Twitter Analytics, etc. While you may only look at a handful of tools, your team overall is probably inundated with data that we’re only half using. It’s not the end of the world to not leverage data as best as it could, but what if you could find potential SEO content opportunities that PPC is spending over $165k on that could be a waste. That’d be worth reading on about, right?
Evaluate the Data You Already Have
To get started on connecting utilizing other teams’ data, you have to pay attention to what the other teams are doing. As digital marketers, you probably have a few clients or projects on your plate, competing priorities, a full inbox, and a limited amount of hours in a day. So, it’s easy to tune out once you’ve gone through your portion of a client call and get back to answering emails. No harm, no foul, right?
If you’re doing this (I definitely have), you’re missing the key to integration: awareness. If you don’t even know what your counterpart PPC and Analytics teams are doing, you won’t be able to understand a client’s full business, develop a holistic strategy, and, for the sake of this post, recycle the data they have.
To find opportunities for recycled data, you’ll need to work collaboratively with all teams on the project: SEO, PPC, and Analytics. Start by asking each team to create a list of all the data sources they regularly access and the intent of each. Once that’s completed for each team, set up a meeting to clarify any questions and brainstorm ideas to recycle the data for other projects. You can even come to the meeting with problems you’ve been trying to solve and see if any of the available data can help.
For this post, we’re going to focus on opportunities to use “recycled” data for SEO strategy. This isn’t meant to serve as a “how to” of the process (if you’re interested, leave a comment below!), but as a thought starter to, well, get you started.
The Power of Search Queries for SEO
WHAT IS IT?: Search queries are the exact terms users searched before they clicked on a paid search ad. Reports on search query performance are available in both Google Adwords and Bing Ads and include information like the exact search query, campaign information, click data, conversion data, etc. If you’re an SEO, this is similar to the organic keyword reports back in 2008 before “not provided” began.
These are used by paid search teams to find irrelevant keywords to negate, uncover themes that need to be improved, or discover queries that perform well that we should purposefully target in campaigns.
WHY DOES IT MATTER?: You do keyword research for any SEO or paid search campaign, right? Why? Because you want to know what people are looking for, so you can actively target their searches and be present with the information, product, or service they need.
Search queries give you the exact queries people search into search engines. What more of an indicator of what someone needs then getting the data right from the source: the user. At Seer, we spend hours on SEO projects conducting audience research into the motivations, needs, questions, and concerns of people as they go through their journey to purchase our clients’ products and services. We’re now using search queries (if the client runs paid search) to get a better understanding behavior before we start interviews to get early insights faster.
HOW DOES IT WORK?: To analyze search queries for SEO, there are four steps in the process:
- Export the Search Query report
- Prepare the search queries by extracting modifiers and organizing into ngrams
- Visualize the data in PowerBI
- Analyze the search terms, modifiers, and ngrams to find insights into the top words people use, their considerations when searching, information on themselves, and more
For one of Seer’s clients, Fender, we did this analysis to get a better understanding of the words people were using to search for their teaching app, Fender Play, to identify new content opportunities.
Uncover Which Modifiers Matters
For years, the adage “Content is king” rang through the halls of agencies and the ballrooms of conferences. I’m not going to get sidetracked into a post on how oversimplified that is, but I’d like to offer an alternative thought that’s not as catchy: “Words matter.” If you haven’t immediately exited this page (thank you for sticking around), I’d like to clarify by saying that for the majority of my 9+ years in digital marketing, I didn’t really consider the importance of the exact words in the content I recommended for clients–except in the keyword sense.
Now, as I’m mostly older and a little wiser, I’ve learned that the words people type into search bars are critical to understand what they need–and it’s just as critical to make sure we’re using them to speak the “language” of our audience.
For Fender, these were the top modifiers of the search terms:
Since these are the words people used most when searching, it’s pretty clear that these terms mattered a lot to them, so we should be including them not only as keywords in title tags, but as part of our CTAs in meta descriptions, page headers, and body copy, and more. If we know that people searched for ‘easy’ over ‘simple,’ it’s a no-brainer which one we should use in our limited meta description space to get folks to click on our listing vs a competitor’s.
Find New Content Opportunities
Search queries allow you to find new content opportunities you may not have discovered before. While there are a lot of methods of content ideation, this one uses “indicator terms” as a way to gain insight into specific considerations people have when search that could indicate new content needs.
An “indicator word” is commonly is a word that specifies an explicit consideration when someone searches. For example, “for” is an indicator word: for parents, for students, for men, etc. It indicates a consideration of the searcher that can be targeted in both SEO and PPC.
Using PowerBI’s extracting options, I created a new column to capture the text after the delimiter “for” and sorted by top unigrams:
It’s easy to see that “beginners” and “beginner” stick out from the list, so it’s a clear content opportunity for Fender if they don’t have content on it already to help this particular audience.
Another call-out is for “kids;” similar to the last example, these users explicitly indicated the audience to which they’re searching for in their query, making it easy to target them with content targeted to their unique needs.
Bonus Time: Looking for PPC Wasted Spend
By using the scatterplot visual where count of Adjectives (or modifiers–not all words indicated are true adjectives) is the x-axis and CPA is on the y-axis, I was able to find adjectives with a high CPA that could indicate wasted spend.
Digging into the second-from-the-top bubble for the modifier “my,” the CPA is $850–way more than the baseline CPA of $53. Here’s what I found when digging into performance:
I don’t know the “I Don’t Know My Name” ukelele song, but I can tell it’s popular and, for some reason, our PPC ad is visible for it. However, I do know enough after research (and a touch of common sense) that people searching how to play a specific song get video results front-and-center to teach them exactly what they’re looking for. They’re probably not going to download a full, paid app just to learn one song they could find on YouTube. Overall, “my” search queries cost almost $1700. They got two conversions, but most of the costs were for the song tutorial. Even if those terms wasted $1200, that could be a nice payment to a content writer to start writing the “for beginners” content we found above.
Mapping Out Content Improvements Based on User Behavior
WHAT IS IT?: Going to the Analytics side of the house this time, our team uses heatmapping for certain analyses to find areas of friction on client’s site.
WHY DOES IT MATTER?: Heatmaps are great for finding areas of a website that elicit low engagement, potentially indicating an opportunity for CRO (Conversion Rate Optimization). Usually, there’s particular interest paid to CTAs and buttons within a heatmap to determine if users are or aren’t taking the next step marketers outline in their journey.
However, this data can also be used to understand what content is or isn’t resonating on a given page to inform content audit recommendations. Instead of just reviewing the actions on a page, we can use heatmaps and mouse tracking to understand what content users are engaging with more or less to see what they’re caring about.
HOW DOES IT WORK?: There are a lot of heatmapping platforms out there, including CrazyEgg, Inspectlet, and ClickTale, but we’ve used HotJar at Seer most often. The specific heatmapping options depend on the tool, but generally we use classic heatmaps and what HotJar calls “move maps” (similar to mouse tracking maps) to understand what content is and isn’t being engaged with and how that can inform our strategy.
The process is simple: once you have your platform of choice installed on the website, begin heatmap and move map tracking for the pages you want to investigate. For this tactic, I’d recommend using it on important landing pages (no surprise here), but, more interestingly, pages that will follow a standard format, like Services pages or a multi-series blog post. For pages that’ll follow this type of format, it’s important to add a heatmap to the first page in the series to see what content is engaged with vs not, so updates can be made in the template before all the other pages in the category go live.
For example, say you’re a Saas company, and you’ve decided to launch Industry pages to speak to the benefits your tool can provide to a variety of different verticals. After audience, keyword, and competitor research, you’ve decided that there will be a dedicated page for each industry that’ll include:
- Overview of the Tool
- Industry-Specific Benefits and Use Cases
- Top Features for Each Industry
- Existing Client Logos
- Client Reviews
Analytics may add heatmap tracking to the page to understand why users are clicking one CTA over another, but SEO teams can utilize this data to look at what content sections are most engaged with. For example, the second section could have a lot of move map activity, but then it drops off for Top Features, then is significantly higher for Existing Client Logos and Client Reviews.
Based on this, we may want to test moving Existing Client Logos and Reviews higher on the page to give users what they’re looking for.
We did this a year ago for Seer’s Career page as an experiment and found that one of our elevated call-out boxes describing our $1k Kaizen budget for all Seer team members had less engagement that a standard paragraph about our compensation strategy. With that insight, I recommended elevating information about our compensation strategy into the call-out since users cared more about it.
The next time you’ll be working on content that will follow a standard template, ask your Analytics team what tracking they plan on adding and look out for an opportunity to recycle this data for SEO improvements.
Living Into Cross-Divisional Integration
If you work with different channel teams in your day-to-day, it’s easy to become complacent in your own world with your own data. By taking a look at what you have access to as a team, you’ll be able to get outside of your typical resources and can find recycled data opportunities to use for your client–without having to request more from your already strapped-for-time POC.
I’ll be continuing this series with posts on how to find similar opportunities for PPC and Analytics teams, so subscribe to our newsletter to be notified when they’re posted.
Do you have any examples of how you’ve used recycled data from other teams or channels? Please add them in the comments below–I’d love to discuss them. Don’t worry, I won’t steal them for the other posts in this series (unless they’re really good and I get your permission 😊).
- UK data chief demands overhaul of campaign rules after Brexit
- Ireland is chasing to keep pace with data analytics revolution
- The need for business analytics is getting a boost from Artificial Intelligence and Internet of Things
- Retrained staff 'key to exploiting big data'
- Read This Before Buying Verisk Analytics, Inc. (NASDAQ:VRSK) Shares
- Belfast-based Analytics Engines secures new partnership with London's National Gallery
- Junkyard Britain? Data shows fly-tipping waste mountains are getting bigger
- New Frontier Data Acquires Zefyr In $10M Deal
- Portugal's declining waste recycling performance increses need for landfill
- Romania ranks last in the EU for battery recycling